the video head impulse test: administering and...

18th Workshop on Neurotology and Medical Audiology

Kolkata, India

Nov 29-Dec 1, 2019

Kamran Barin, Ph.D.

Assistant Professor, Emeritus

Department of Otolaryngology

The Ohio State University

barin.1@osu.eduDisclosure: Consultant to Interacoustics and Bertec Corp.

The Video Head Impulse Test:

Administering and Interpreting the Test

2

Overview

• Video head impulse test (vHIT) has become an important part of

evaluation for patients with dizziness and other balance

disorders

• The test is simple to perform and provides valuable information

about all six semicircular canals

• Two main parameters, VOR gain and catch-up saccades, are

used for interpretation of the test results but many questions

remain:

– What is the clinical significance of VOR gain?

– What is the clinical significance of overt and covert catch-up saccades?

– How to avoid artifacts and identify them when they occur

3

Bedside HIT – Normal Responses

• HIT consists of moving the head

using small-amplitude high-velocity

high-acceleration unexpected head

movements as the patient fixates on

a stationary target straight ahead

– In order to avoid contamination from the

oculomotor pathways, head velocities

should be above 100°/sec for lateral

impulses and above 50°/sec for vertical

impulses

• In normal subjects, VOR eye

velocities match the head impulse

velocities in the opposite direction

allowing the eyes to remain on the

target after an impulse

4

vHIT – Normal Responses

• Head impulses toward the

canal cause excitation from that

canal

– Changes in the neural firing are

proportional to the head velocity

• Head impulses away from the

canal cause inhibition from that

canal

– Neural firing is clipped

(saturates) at 0 spikes/sec and

does not provide an accurate

measure of head velocity

• Both labyrinths contribute to

generating eye movements

– The contribution of the

ipsilateral labyrinth is larger

(Ewald’s second and third laws)

5

vHIT – Normal Responses

• In normal subjects, the VOR gain is around 1 at lower head

velocities but begin to decline slightly at higher head velocities

• At higher head velocities (~300°/sec) even normal subjects may

have catch-up saccades

– Stay below 250°/sec for lateral head impulses and below 200°/sec for

vertical head impulses

6

Bedside HIT – Responses in Unilateral Lesions

• In patients with vestibular lesions, the

eyes fall short of the target for head

impulses toward the damaged side

• After the head comes to a stop, a

catch-up saccade is generated to

reach the target

– These saccades are called overt

saccades because they are visible to the

naked eye

X X

7

vHIT – Responses in Unilateral Vestibular Lesions

• For head impulses toward the

side of lesion, the neural

response from the damaged

side is reduced or abolished

• The neural response from the

intact side is saturated and no

longer proportional to head

velocity

• The resulting eye velocity does

not match head velocity and the

eyes fall short of target

• VOR Gain = Eye Move./Head

Move. << 1 (decreases rapidly

with increasing head velocity)

8

vHIT – Responses in Unilateral Vestibular Lesions

• For head impulses away from

the side of lesion, the neural

response from the intact side is

proportional to head velocity

• The neural response from the

damaged side is again reduced

or abolished

• The resulting eye velocity is

closer but still does not match

head velocity and the eyes fall

somewhat short of target

• VOR Gain = Eye Move./Head

Move. < 1 (decreases with

increasing head velocity but not

as rapidly as the VOR gain for

head impulses toward the side

of lesion)

9

vHIT – Catch-Up Saccades

• The mechanism for triggering catch-up

saccades is intuitively understood to be the

difference between the gaze and target

positions at the end of a head impulse

• After the position difference is detected and

the saccade is initiated, it takes about 80-100

ms for the eyes to begin to move

• These saccades have long latencies (>~250

ms)

– They are called overt saccades because they

occur after the head movement and are visible

to the naked eye

10

Bedside HIT – Responses in Unilateral Lesions

• Some patients are able to initiate

saccades during head movements

– The saccades are called covert saccades

because they often (but not always!)

occur during head movements and are

not visible to the naked eye during the

bedside HIT

• These saccades have short latencies (<~200

ms)

X X

11

vHIT – Covert Saccades

• Coverts saccades require some form of

learning and prediction

• Coverts saccades usually are followed by a

small overt saccade

• In preliminary reports, covert saccades have

been associated with compensation

– Better dynamic visual acuity, improved

balance, and reduced symptoms

12

vHIT Summary – Unilateral Lesions

• VOR gain is << 1 for head impulses toward the side of lesion and declines with

increasing head velocity

• VOR gain is < 1 for head impulses away from the side of lesion but does not decline

as much with declining head velocity

• Significant catch-up saccades are present for impulses toward the side of lesion

• Catch-up saccades can be present for impulses away from the side of lesion also but

they are not as large and start at higher head velocities

• Initially, most of the catch-up saccades are the long-latency type (overt)

• Over time, with learning and prediction, short-latency (covert) saccades may develop

13

vHIT Summary – Unilateral Lesions

• In acute lesions, fast phases of spontaneous nystagmus will be intermixed with catch-

up saccades for impulses toward the side of lesion and in the opposite direction of

typical catch-up saccades for impulses away from the side of lesion

• Spikes for spontaneous nystagmus can occur before or after head impulses

Interpretation of vHIT

15

Different Patterns of Valid vHIT Results

Normal

UW Compensated

UW Acute

BW Partial

16

vHIT Interpretation – Step-by-Step Guide

• Different types of eye movements in vHIT

– Slow VOR eye movements

• VOR eye velocities are either shaped like the head velocities or in case of severe loss,

they will appear as the clipped version of the head velocities

– Fast eye movements

• Long-latency catch-up saccades (overt)

• Short-latency catch-up saccades (covert)

• Fast phases of spontaneous nystagmus

– All others (artifacts)

17

vHIT Interpretation – Step-by-Step Guide

• Step 1 – Identify the artifacts and determine if the test is interpretable

– Delete impulses with artifacts as you need only a few impulses for interpretation

• When trying to identify saccade types, look at the individual impulse tracings

18

vHIT – Artifacts and Saccade Look-Alikes

• Two consecutive saccades going in

opposite directions at about 80-100 ms

apart

– The patient is looking around

– Reinstruct the patient

• Biphasic or uniphasic artifacts due to

eye blinks, eyelids obstructing part of

the pupil, or LED light intruding on the

pupil

– Shine a light to shrink the pupil

– Pay attention to the display while

performing the test

– Delete affected impulses

– Recording the video of eye movements

can help with identifying these artifacts

19

vHIT – Artifacts and Saccade Look-Alikes

• High-frequency oscillations

due to pupil detection issues

– Adjust the cameras, the focus,

and the threshold (in some

systems)

bad calibration

bumpslippage

Abnormally high or low VOR gain in the absence of

catch-up saccades usually indicates an artifact (do a

quick saccade test to rule out saccadic palsy)

Gain ~ 0.4

Gain ~ 0.4

20

vHIT – Other Artifacts

21

vHIT – Spontaneous Nystagmus

• Step 2 – Account for spontaneous

nystagmus

– Fast phases of spontaneous

nystagmus appear as spikes in eye

velocity tracings

– Spontaneous nystagmus fast

phases can occur before or after

head impulses

– For typical spontaneous nystagmus

that beats away from the side of

lesion, spikes appear in the

opposite direction of VOR eye

movements following head impulses

toward the intact side

– Fast phases of spontaneous

nystagmus are intermixed with the

catch-up saccades during head

impulses to the lesion side

Modifying the display parameter can help

with identifying nystagmus fast phases

22

vHIT Interpretation – Step-by-Step Guide

• Step 3 – Determine if abnormal catch-up saccades are present

– Identify true catch-up saccades using the latency of first saccade

• Short latency ~80-225 ms, long latency ~225-350 ms

– Determine significant saccades using the saccade peak velocity

• Use peak saccade velocity > 100⁰/sec or > half of the peak head velocity

• Velocity may be too variable but there is no other established criteria

• When there are no significant catch-up saccades, check the VOR gains and

if they are within normal limits (close to 1), vHIT should be considered within

normal limits

• If VOR gains are not within normal limits in the absence of abnormal catch-

up saccades, consider presence of an artifact

23

vHIT Interpretation – Normal Results

• No clearly-identifiable catch-up saccades

• VOR gains are close to 1 bilaterally

– Head velocities are approximately equal for right-left impulses and within the

optimal range

– No other sign of artifacts

24

vHIT Interpretation – Normal Results

• There are few catch-up saccades bilaterally for higher head velocities

− Saccade velocities are considerably smaller than the corresponding head

velocities

25

vHIT Interpretation – Step-by-Step Guide (continued)

• Step 4 – If abnormal catch-up saccades are present in one direction only

and the VOR gain is abnormal for the same side, consider vHIT consistent

with a unilateral lesion

– If the VOR gain is normal in the presence of abnormal saccade, still consider the

vHIT abnormal (may signify mild lesions)

– The VOR gain to the opposite direction is usually normal but it does not

contribute to the interpretation even if it is abnormal

• If abnormal catch-up saccades are present to both directions, there is no

easy way to distinguish between purely unilateral and partial bilateral lesions

– Assume unilateral unless proven otherwise

– For unilateral lesions, catch-up saccades are more frequent to the side of lesion

and start at lower head velocities

– For unilateral lesions, VOR eye velocities are clipped/saturated for one direction

but proportional to head velocities in the opposite direction

– For bilateral lesions, the sum of right and left VOR gains is less than one

26

vHIT Interpretation – Unilateral Vestibular Deficit

• Abnormal catch-up saccades to both directions but more significant for

rightward head impulses

• Saturated VOR eye velocities for rightward head impulses and proportional VOR

velocities for leftward head impulses denote a right unilateral vestibular deficit

– VOR gain for leftward head impulses may be abnormal but interpretation is the

same

27

vHIT Interpretation – Bilateral Vestibular Deficit

• Complete bilateral loss – Abnormal catch-up saccades and near 0 VOR gain

for both directions

28

Summary

• Head velocities must be within a specific range and right-left velocity profiles

must be approximately the same for valid interpretation of vHIT

• Catch-up saccades can be helpful in identifying abnormal vHIT

– Abnormal catch-up saccades are more frequent and start occurring at lower head

velocities

– Presence of short-latency catch-up saccades has been associated with improved

dynamic visual acuity, better balance, and reduced symptoms but more work is

needed

• In the presence of abnormal catch-up saccades, vHIT should be considered

abnormal regardless of whether the VOR gain is abnormal or not

– Presence of consistent catch-up saccades with normal VOR gains is likely to

represent a mild lesion

• Conversely, abnormal VOR gain in the absence of catch-up saccades should

be investigated further for possible artifacts

29

Let’s Do a Test!

• Can we estimate the loss of canal function in this vHIT?

• Is the loss confined to the right canal?

R Gain = 0.42

L Gain = 0.87

Total R-L Gain = 1.29

Mean R-L Gain = 0.645

30

VOR Gain and the Level of Vestibular Loss

R Gain = 0.42

L Gain = 0.87

Total R-L Gain = 1.29

Total Loss% =

(2 – 2 x Total R-L VOR Gain)*100 =

(2 – 1.29)*100 = 71%

DO NOT DO THIS!

(Different frequency ranges)

|UW%| = %

= 55%

1 – Mean VOR Gain

Mean VOR Gain

31

vHIT – Clinical Applications

• For the first time, isolated abnormalities in vertical canals and their afferent

pathways can be identified

– In vestibular nerve abnormalities, vHIT can determine which branches of

vestibular nerve are involved and can determine when and if function returns to

the vestibular nerve

• In the case of acute vertigo, can differentiate between cerebellar strokes and

peripheral vestibular lesions (Newman-Toker et al, 2013)

• Can be used for serial testing (e.g., monitoring Gentamycin therapy for

Meniere’s or monitoring vestibulotoxicity of different agents)

• Can be used in place of rotation testing in patients with bilateral caloric

weakness

• Can be modified for testing children (e.g., before cochlear implant)

• Cost-effective because it reduces the need for unnecessary tests